coreweave-fabric-diagnostics

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The required workflow instructs the agent to ingest and process free text pasted by the user (such as NCCL logs, pod specs, and diagnostic output) via a deterministic script, but since user-submitted text comes from the direct user interaction rather than an unauthenticated or monitored external channel like a queue or feed, it falls under low/active ingestion or no untrusted threat.